Search
Browse
Create
Log in
Sign up
Log in
Sign up
Upgrade to remove ads
Only $2.99/month
Research Methods - Quiz 2 - STATS
STUDY
Flashcards
Learn
Write
Spell
Test
PLAY
Match
Gravity
Terms in this set (45)
Null Hypothesis
Statement regarding the value of unknown parameters. Typically will imply no association between explanatory and response variables in our applications.
Alternative Hypothesis
Statement contradictory to the null hypothesis (will always contain an inequality)
Type 1 Error
Rejecting of a true null hypothesis
i.e. when a jury convicts an innocent person
i.e. accepting an alternative hypothesis when there really isn't enough evidence
0.05/0.01
P (Type 1 error) is approximately how much in terms of value?
Type 2 Error
Accepting (do not reject) a false null hypothesis.
i.e. accepting a null hypothesis even though there is evidence that is likely to support a research or alternative hypothesis.
i.e. when a guilty person is acquitted.
Alpha
Symbol for probability of a Type 1 error is a(n)
Inversely related with the value of the Type 2 error, meaning decreasing the value in one increases the value in the other.
Beta
Symbol for probability of a Type 2 error is a(n)
Inversely related with the value of the Type 1 error, meaning decreasing the value in one increases the value in the other.
reduces
Increased value in T1 error (alpha) ___________ the value of a T2 error (beta)
increases
Decreased value in T1 error (alpha) ________ the value of a T2 error (beta)
reject
If p is less than the alpha value then we __________ (fail to reject or reject) the null hypothesis
fail to reject
If p is equal to or more than the alpha value then we (fail to reject or reject) the null hypothesis
fail to reject = cannot say we accept
alpha
___________ value defines how much of an error we are willing to make to say we made a difference in our experimental manipulation.
2-Tailed Test
The critical value is the number that separates from the "blue zone" from the middle (+/- 1.96)
In a t-test, in order to be statistically significant, the t-score needs to in a blue zone t= +/- 1.96 from 0
1-Tailed Test
The critical value is either + or -, but not both.
1-Tailed Test
When statistical significance (p <0.05) is greater than or equal to 1.645, is an example of a
2-Tailed Test
When statistical significance (p<0.05) is: 1) greater than or equal to 1.645 and 2) less than or equal to -1.645, is an example of a
Descriptive Statistics
Describe characteristics, organize, summarize, condense data
Inferential Statistics
Test hypothesis, make conclusions, interpret data, understand relations
Continuous/Interval/Ratio
Equal Increments
Ordinal/Rank
In order but not equal intervals
Categorial (Nominal)
Names
Statistics
These are calculated using a sample data.
Parameters
These are characteristics of a population data.
Kurtosis/Skewness
Shape of data is measured by...
Skewness
Measures of asymmetry of data
Right Skewed (positive)
Longer right tail
Left skewed (negative)
Longer left tail
Kurtosis
Measures peakedness of distribution of data.
General forms of this:
-leptokurtic (sharp, high peak)
-mesokurtic (normal peak)
-platykurtic (flat. low peak)
Homogeneity
__________ (homogeneity or heterogeneity) of variances, meaning the data from groups have similar variance.
Independent samples t-test
Use this type of test if comparing 2 groups (e.g. autism/control)
Paired samples t-test
Use this type of test if comparing 2 conditions within the same group (not independent observations)
ANOVA
Use this type of test if comparing >2 groups, >conditions, or both within and between groups
Pearson R-correlation
Use this type of test if measuring association between 2 variables.
Multiple Regression
Use this type of test if measuring between >2 variables.
ANOVA
This test is an extension of the t-test and it is used for two or more groups where the difference between means are tested for normally distributed variables
One-Way ANOVA
(Between groups), this test is similar, if not the same, as the independent samples t-test by finding the following:
-Normally distributed dependent variable (outcomes)
-Equal variances between the groups
-Groups are independent
5
Error rate of making a type 1 error for each test is up to _________ (%)
Bonferroni correction
This is the most conservative adjustment of "p" when correcting MULTIPLE COMPARISONS post-hoc.
Assuming all tests independent, divide "p" by the number of tests
Turkey, Sheffe, FDR
All three methods alter critical p-value to account for the fact that you've done multiple comparisons, but are less conservative than Bonferroni.
Nonparametric Test Procedures
These tests do not involve population parameters (e.g. probability distributions)
-Ratio or interval
-Ordinal
-Nominal
Disadvantage
The following are _________ (adv/disadv) of nonparametric tests:
-May waste information (e.g. using parametric procedures, throwing away distributions of intervals by converting them into an ordinal scale --> scores ranging 1-100 to replace with 3 categories of low, medium, and high.
Advantage
The following are _________ (adv/disadv) of nonparametric tests:
-Used with all scales
-Easier to compute
-Make fewer assumptions
-Need not involve population parameter
-Results may be exact as parametric procedures.
Nonparametric Test Types
-Sign Test
-Wilcoxon Rank Sum Test
-Wilcoxon Signed Rank Test
-Kruskal Wallis H-test
-Speakman's Rank Correlation Coefficient
Categorical Data
This is called a test of frequency -- how often something is good
(aka: goodness of fit test, test of homogeneity)
Chi Square
Can be used when finding M:F ratio in Autism vs General Population
This test compares observed frequency to what we expected.
-Any number squared is a positive number, therefore, area under the curve starts at 0 and goes to infinity.
-To be statistically significant, the value needs to be in the upper 5% (alpha score = 0.05)
YOU MIGHT ALSO LIKE...
Statistics Chapter 7
17 terms
Stats ch 9
30 terms
Stats 12: Chapter 8- Hypothesis testing for popula…
146 terms
STAT - EXAM 3 - Hypothesis Testing and C…
24 terms
OTHER SETS BY THIS CREATOR
Research Methods Quiz - Statistical and Practical…
29 terms
PEDS LANG EXAM 3: Literate Language/ Discourse
35 terms
PEDS LANG EXAM 3 - NARRATIVES
50 terms
PEDS LANG EXAM 3 - Complex Syntax
24 terms